Journal of the American College of Nutrition, Vol. 17, No. 1, 30-35 (1998)
Published by the American College of Nutrition
Glycemic Index and Insulin Response to a Liquid Nutritional Formula Compared with a Standard Meal
Thomas E. Edes, MD and
Jayendra H. Shah, MD
Department of Medicine, Harry S. Truman Memorial Veterans’ Hospital and University of Missouri, Columbia, Missouri and Department of Medicine, Tucson VA Medical Center and University of Arizona College of Medicine, Tucson, Arizona
Address reprint requests to: Thomas E. Edes, MD, Harry S. Truman Memorial VA Hospital, and University Hospital, 800 Hospital Drive, Columbia, MO 65201
 |
ABSTRACT
|
|---|
Objective: To determine the glycemic index and metabolic responses to a nutritional formula, and to compare these responses to those following an oral glucose meal and a standard test meal.
Methods: Six male and six female healthy non-diabetic volunteers aged 18 to 48 years met screening examination and laboratory assessment criteria. Three test meals were administered, each containing 50 g of carbohydrate: nutritional formula (NF), standard test meal (ST) and a glucose test meal (GT). Each subject underwent the three test meals on separate days in randomized sequence. Blood samples were taken at intervals over 5 hours for determination of glucose, insulin and triglycerides.
Results: The glycemic index was similar for the NF (60.8±13.1) and for the ST (57.8±12.9) meals. The incremental area under the curve for glucose was similar for NF and ST, but each was significantly lower than for the GT meal. The total area under the curve for insulin was significantly greater for the NF meal than for the ST meal. The serum triglyceride responses were similar for NF and ST meals.
Conclusion: In healthy non-diabetic subjects, the blood glucose and triglyceride responses are similar for a nutritional formula compared to an isoenergetic standard test meal. However, the insulin response differs. This information is important in managing tube-fed patients.
Key words: glycemic index, nutritional formula, nutrition support, insulin, immunoreactive insulin
Abbreviations: AUCG=area under the curve for glucose • AUCI=area under the curve for insulin • GT=glucose test meal • IDDM=insulin dependent diabetes mellitus • IRI=immunoreactive insulin • NIDDM=noninsulin dependent diabetes mellitus • NF=nutritional formula • ST=standard test meal
|
INTRODUCTION
|
|---|
The metabolic response to carbohydrate ingestion plays an important role in health and in disease states such as diabetes mellitus and hyperlipidemia. The glycemic index (GI) quantifies the blood glucose change after eating a certain food compared to the change after eating a similar amount of carbohydrate as glucose [1,2]. Different carbohydrates can have a significantly different GI [3]. The GI has potential value in the management of persons with diabetes mellitus. Their levels of fasting blood glucose, hemoglobin A1C and urinary C-Peptide are lower when on a low GI diet than when on a high GI diet [4,5]. The GI for a specific food is similar in subjects with non-insulin dependent diabetes mellitus (NIDDM) as in those with insulin dependent diabetes mellitus (IDDM) [6]. In children with IDDM, a higher insulin dose is required to maintain blood glucose while on a high GI diet compared to when on a low GI diet [7].
Although the GI is a specific indicator of blood glucose response, glucose is not the only parameter which may be affected by the GI of a food. Serum triglycerides and cholesterol are modestly decreased by a low GI diet, in nondiabetics as well as those with NIDDM [8,9].
Liquid nutritional formulas are used extensively. They may be used to supplement the intake of those who do not voluntarily consume adequate nutrition or who would benefit from weight gain. They may be used as the sole nutritional intake for a few days or for years in individuals who are unable to swallow food and require tube feeding. Although liquid nutritional formulas are frequently used, their GI and metabolic effects have not been studied in detail. Information on the metabolic response is limited and is not compared with a standard meal [10]. The present study was designed to investigate the GI and metabolic responses to a nutritional formula in normal healthy subjects, compared to the metabolic responses of oral glucose and a standard test meal.
|
MATERIALS AND METHODS
|
|---|
Subjects
The study protocol was approved by the Institutional Review Boards of the University of Missouri-Columbia and the Harry S. Truman Memorial Veterans’ Hospital. Following this approval, healthy volunteers were recruited by local advertisement. The first six female and six male volunteers who fulfilled screening criteria and agreed to the study stipulations were enrolled. The demographic information on these subjects is summarized in Table 1. Screening criteria included healthy male or female subjects between the ages of 18 and 50 years, no known diabetes mellitus nor other chronic illness, no medications, and body mass index between 19 and 30 kg/m2. All potential subjects gave written informed consent. They underwent an initial screening medical history and physical examination, urine drug screen and laboratory assessment. The serum chemistry laboratory assessment included measurement of sodium, potassium, chloride, bicarbonate, blood urea nitrogen, glucose, creatinine, calcium, phosphorous, albumin, total protein, direct bilirubin, total bilirubin, alkaline phosphatase, aspartate transaminase, alanine transaminase, uric acid, cholesterol, lactate dehydrogenase, and creatine phosphokinase. An abbreviated (3-hour) screening Oral Glucose Tolerance Test (OGTT) was then administered using 75 g of glucose in solution. Subjects were excluded for fasting blood glucose above 115 mg/dl, an abnormal screening OGTT, or a clinically significant abnormality on the serum chemistry profile.
Test Meals
Each subject met with our dietitian prior to entering the test meal phase. They were instructed and advised to consume a moderately high (at least 300 g/day) carbohydrate diet for 3 days prior to each test meal.
They were advised to maintain similar exercise and eating habits for 3 days prior to each test meal. Each test meal was separated by at least 4 days. Each of the 12 subjects received each of the three test meals but not in the same sequence. Each subject was randomized to receive one of the six possible sequences of test meals, such that each test meal sequence was administered to two subjects.
The three test meals each contained 50 g of carbohydrate. The glucose test (GT) solution contained 200 kcal, the nutritional formula (NF) contained 367 kcal, and the standard test meal (ST) contained 361 kcal. The powdered nutritional supplement (Enercal Plus, formulated by Wyeth-Ayerst International, Inc.) was administered as 240 cc of formula containing 367 kcal, 14 g protein, 50 g of available carbohydrate, 12 g fat, 48 mg (80% RDA) of ascorbic acid, and 25% RDA of all other known essential vitamins and trace elements. The carbohydrates in NF were maltodextrin (78%), sucrose (20%), and lactose (2%). The ST meal developed by our registered dietitian consisted of two slices of white bread, one and one-half eggs, 1 teaspoon margarine and 6.5 oz of orange juice. This ST meal contained 361 kcal, 13 g protein, 50 g of available carbohydrate, 13 g fat, and 2 g nonavailable carbohydrate fiber.
Each test meal session began between 7:00 and 8:30 a.m. Prior to each session, the subjects were weighed and interviewed. They were allowed to drink water during the 12-hour fast prior to each test meal, but were advised not to consume any product containing caffeine. A urine drug screen was collected prior to each test meal. Subjects who reportedly had not complied with the 12-hour fast, who had taken medication within 3 days prior to the test meal, or who had a potentially interfering substance identified in the urine drug screen were rescheduled for the test meal at a later session. Only one of the 36 test meals was rescheduled because of a potentially interfering substance.
On the morning of the OGTT and each test meal, a 1
inch silastic catheter was placed in an arm vein using aseptic technique. Blood was drawn through the catheter 30 minutes prior to each test meal, then at 30 minute intervals throughout the 5-hour test meal session. Subjects were asked to remain in the area of the testing facility throughout the 5 hours of each session. They were allowed to read, stand occasionally, walk the short distance to the bathroom as needed, or sit in a comfortable chair or on the floor during the testing period.
Serum Analysis
For each test meal, 12 serum samples were obtained at 30-minute intervals beginning 30 minutes prior to the test meal and ending 5 hours after the test meal. Glucose was determined for each sample by a glucose oxidase method. Insulin was measured in each sample by a radioimmunoassay technique [11].
Triglycerides were measured at -30, 0, 60, 120, 180, 240, and 300 minutes by a lipase method on a commercial Cobas-Bio Analyzer.
Statistical Analysis
Comparisons between the groups were made by two-way analysis of variance. The GI was determined according to the following formula: GI=(area under the curve for glucose response in given test meal)÷(area under the curve for the glucose response in glucose meal)x100, where the test meal and glucose meal each contained the same quantity (50 g) of carbohydrate. The area under the curve was determine as the area of those increments above baseline, with the baseline being the average of values at -30 and 0 minutes. The GI is expressed as percent. GI data were analyzed using a non-parametric approach for the 2x2 crossover design. This approach was also used to analyze the Insulin Index, the serum immunoreactive insulin responses over time after each test meal. The paired t-test was used to compare test meal results at any given time.
|
RESULTS
|
|---|
Demographic characteristics of volunteers: Six males and six females completed all three test meals (Table 1). Their age range was 18 to 48 years, their body mass index ranged from 21 to 29, and their anthropometric measurements fell within acceptable norms for health.
Serum Glucose and Incremental Area Under the Curve for Glucose
The results of serum glucose analysis expressed as mean±SEM are shown in Fig. 1. Combining data from all subjects, peak glucose levels occurred 30 minutes after each test meal as well as the screening OGTT. The serum glucose then fell to baseline levels at 90 minutes for NF and ST meals, but at 120 minutes for the GT meal. Glucose levels then remained at approximately their baseline for the remainder of the 5 hours. At 30, 60 and 90 minutes, serum glucose was lower for the NF meal than for the GT meal (p<0.03), and was lower for the ST meal that for the GT meal (p<0.01). Serum glucose levels for the NF meal were not significantly different from the ST meal at any time. If the glucose data are separated into male and female categories, the mean glucose levels are consistently but not significantly higher in the males compared to the females at baseline, 30, 60 and 90 minutes (Figs. 2 and 3).
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 1. Serum Glucose Response
All Subjects (N=12)
Values are glucose in mg/dL expressed as mean±SEM. At 30, 60 and 90 minutes, p(NF vs. GT)<0.03 and p(ST vs. GT)<0.01.
OGTT=oral glucose tolerance test (75 g glucose), NF=nutritional formula, ST=standard test meal, GT=glucose test (50 g glucose).
|
|
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 2. Serum Glucose Response
Females (N=6)
Values are glucose in mg/dL expressed as mean±SEM.
OGTT=oral glucose tolerance test (75 g glucose, NF=nutritional formula, ST=standard test meal, GT=glucose test (50 g glucose).
|
|
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 3. Serum Glucose Response
Males (N=6)
Values are glucose in mg/dL expressed as mean±SEM.
OGTT=oral glucose tolerance test (75 g glucose), NF=nutritional formula, ST=standard test meal, GT=glucose test (50 g glucose).
|
|
The incremental area under the curve for glucose (AUCG) for each of the three test meals was determined. The AUCG for the NF meal was similar to that observed for the ST meal. However, the AUCG for ST as well as for NF were significantly lower than the AUCG for the GT test meal (p<0.05) (Table 2). These results were similar whether males and females were evaluated separately or combined as a group. The responses in males compared to females were not significantly different.
Glycemic Index
For all 12 subjects (Table 2), the GI for the NF meal (60.8±13.1) was similar to that for the ST meal (57.8±12.9). If male and female data were analyzed separately, there remained no significant difference in GI between the NF meal and the ST meal.
We also compared the GI of females and males. The GI was greater in females (69.3±26.5) than in males (52.3±5.2) for the NF meals; the difference was not statistically significant. A similar result was obtained for the ST meal, with females (66.5±23.9) greater than males (49.1±11.3) but again not statistically different.
Insulin
Serum immunoreactive insulin (IRI) was measured at 30-minute intervals. IRI levels were low before each test meal, peaked at 30 minutes after each test meal, and returned to baseline by 210 minutes after each test meal (Fig. 4). Although the time course pattern was similar for each test meal, the insulin level was significantly higher for the NF meal than for the ST meal over the time interval 30 through 180 minutes. Separating the data into male and female groups, the trends were the same for peak insulin at 30 minutes and return to baseline by 210 minutes, but in females the insulin levels for the NF meal were not significantly different than those observed for the ST meal.
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 4. Insulin Response
All Subjects (N=12)
Serum Immunoreactive Insulin (IRI) levels in McU/ml expressed as mean±SEM.
OGTT=oral glucose tolerance (75 g glucose), NF=nutritional formula, ST=standard test meal, GT=glucose test (50 g glucose).
|
|
The insulin index for each test meal, expressed as percent compared to GT meal, was significantly higher (p<0.02) for the NF meal (149±18) than for the ST meal (91±11) (Table 2).
The incremental area under the curve for insulin (AUCI) is probably the most representative indicator of total insulin release. In response to a meal, the AUCI was significantly higher for the NF meal than for the ST meal (Table 2) for all subjects combined and for males alone, but not for females alone.
Triglycerides
Serum triglycerides did not change over the 5 hours following the GT meal. In both the NF meal and ST meal there was a trend of increasing TG levels to 120 minutes with a decline from 180 to 300 minutes, but these changes were relatively small (Fig. 5) and not statistically significant.
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 5. Serum Triglyceride Response
All Subjects (N=12)
Serum triglycerides expressed in mg/dL as mean±SEM.
OGTT=oral glucose tolerance (75 g glucose), NF=nutritional formula, ST=standard test meal, GT=glucose test (50 g glucose).
|
|
|
DISCUSSION
|
|---|
The primary finding of this study indicates that in healthy males and females, the GI for the nutritional formula is not different from the GI for an isocaloric standard test meal. Since the GI of a food affects diabetic control and insulin requirements and the GI for a food in a nondiabetic person correlates with the GI for that food in a diabetic patient [12], this information on feeding formula may be useful in the management of persons with diabetes. Since the GI of the individual components of a given meal may [13] or may not [14] predict the actual GI of the meal, having information on the GI for a complete tube-feeding formula is clinically valuable.
In addition, the pattern of blood glucose rise and return to baseline was nearly identical for the NF meal and the ST meal. The incremental area under the curve for glucose was similar for both the NF meal and the ST meal, while this value was lower for each of these test meals than for the GT meal. Assessing the metabolic response collectively by the glucose curve, the GI and the AUCG, one finds a consistent similarity between the NF meal and the ST meal. This similarity suggests that the rate of carbohydrate absorption and conversion to glucose is nearly identical for the NF and ST meals.
The rate of absorption of a food appears to be the primary factor in determining the glycemic index [15]. Rate of absorption of food is affected by its fiber content and viscosity and the timing of food intake [8]. One might expect that a liquid nutritional formula with no fiber, ingested over approximately a 5 to 10 minute period, would have a greater glucose response than an isoenergic, isocarbohydrate standard meal. However, it is becoming increasingly apparent that the GI of a food cannot necessarily be predicted by its carbohydrate content [16]. Isoenergic, isocarbohydrate feedings of white bread or white rice yield significantly different blood glucose responses [17]. Fiber content [18,19] and even the vitamin and mineral content of a food may affect the blood glucose response [20].
Although the blood glucose response was similar for the NF meal and the ST meal, the insulin response differed. The timing and pattern of insulin levels was similar, but the levels were higher after the NF meal than after the ST meal. Since insulin release is affected by numerous factors including certain amino acids, protein content [21], fat content [22], and types of carbohydrate, and our ST meal and NF were not identical in content, we cannot provide a specific explanation for the difference in insulin release between the NF and ST meals. Whether this difference in insulin release in healthy, nondiabetic subjects would be reflected in a difference in insulin requirements or in blood glucose response in diabetic subjects remains unknown. A difference in insulin release, independent of blood glucose homeostasis, may have an impact on serum lipids [15].
In addition to the potential value of GI in the management of disorders of blood glucose and lipids, there may be other applications. Nutritional formulas, both liquid and powder, are used in compromised hospitalized patients, as well as in healthy individuals to promote weight gain and enhance the outcome of physical training. A higher GI of foods consumed after exercise promotes recovery of muscle glycogen stores [23] and reduces muscle fatigue [24]; a lower GI of foods consumed before exercise enhances endurance [25]. Further investigations are warranted to determine metabolic response of nutritional formulas in NIDDM and IDDM subjects, as well as for potential applications in athletic performance.
|
CONCLUSION
|
|---|
In healthy nondiabetic subjects, the blood glucose and triglyceride responses were similar for this nutritional formula compared to an isoenergetic standard test meal. However, the insulin response differed. While these responses may not be the same for formulas or meals of different nutrient composition than were used in this study, this information is important in managing tube-fed patients, and may have application in high-level athletic performance.
|
APPENDIX
|
|---|
Enercal Plus Nutritional Information
|
ACKNOWLEDGMENTS
|
|---|
This study was supported by a grant from Wyeth-Ayerst International, Incorporated, Philadelphia, PA. Support was also received from the Department of Veterans Affairs and the School of Medicine, University of Missouri-Columbia. The nutritional formula used in this study was Enercal Plus, manufactured by Wyeth-Ayerst International, but not marketed in the United States. The authors gratefully acknowledge the excellent technical assistance of Jodie L. McIntire and Sudhir Batchu, the dietitian assistance of Shelly Nichols, the administrative assistance of Isha Bhattacharyya, PhD, and the statistical assistance of Alan Dugdale, MD.
Received February 1, 1997.
Revised July 1, 1997.
Accepted July 1, 1997.
|
REFERENCES
|
|---|
- Jenkins DJ, Wolever TM, Taylor RH: Glycemic index of foods: a physiologic basis for carbohydrate exchange.
Am J Clin Nutr
34:
362–366,
1981.[Abstract/Free Full Text]
- Wolever TM, Jenkins DJ, Jenkins AL, Josse RG: The glycemic index: methodology and clinical implications.
Am J Clin Nutr
54(5):
846–854,
1991.[Abstract/Free Full Text]
- Hughes TA, Atchison J, Hazelrig JB, Boshell BR: Glycemic responses in insulin-dependent diabetic patients: Effect of food composition (see comments).
Am J Clin Nutr
49(4):
658–666,
1989.[Abstract/Free Full Text]
- Jenkins DJ, Wolever TM, Buckley G, Lam KY, Giudici S, Kalmusky J, Jenkins AL, Patten RL, Bird J, Wong GS: Low-glycemic-index starchy foods in the diabetic diet.
Am J Clin Nutr
48(2):
248–254,
1988.[Abstract/Free Full Text]
- Brand JC, Colagiuri S, Crossman S, Allen A, Roberts DCK, Truswell AS: Low-glycemic index foods improve long-term glycemic control in NIDDM.
Diabetes Care
14(2):
95–101,
1991.[Abstract]
- Wolever TM, Jenkins DJ, Josse RG, Wong GS, Lee R: The glycemic index: similarity of values derived in insulin-dependent and non-insulin-dependent diabetic patients.
J Am Coll Nutr
6(4):
295–305,
1987.[Abstract]
- Weyman-Daum M, Fort P, Recker B, Lanes R, Lifshitz F: Glycemic response in children with insulin-dependent diabetes mellitus after high- or low-glycemic-index breakfast.
Am J Clin Nutr
46(5):
798–803,
1987.[Abstract/Free Full Text]
- Jenkins DJ, Jenkins AL, Wolever TM, Vuksan V, Rao AV, Thompson LU, Josse RG: Low glycemic index: Lente carbohydrates and physiological effects of altered food frequency.
Am J Clin Nutr
59(3 Suppl)
706S–709S,
1994.[Abstract/Free Full Text]
- Miller JC: Importance of glycemic index in diabetes.
Am J Clin Nutr
59(3 Suppl):
747S–752S,
1994.[Abstract/Free Full Text]
- Peters AL, Davidson MB: Effects of various enteral feeding products on postprandial blood glucose response in patients with type I diabetes.
JPEN
16:
69–74,
1992.[Abstract]
- Herbert V, Law K, Gottlieb GW, Blecher SJ: Coated charcoal immunoassay of insulin.
J Clin Endocrinol Metab
25:
1375–1384,
1965.[Medline]
- Indar-Brown K, Norenberg C, Madar Z: Glycemic and insulinemic responses after ingestion of ethnic foods by NIDDM and healthy subjects.
Am J Clin Nutr
55:
89–95,
1992.[Abstract/Free Full Text]
- Wolever TM, Jenkins DJ, Vuksan V, Josse RG, Wong GS, Jenkins AL: Glycemic Index of foods in individual subjects.
Diabetic Care
13:
126–132,
1990.
- Coulston AM, Hollenbeck CB, Swislocki AL, Reaven GM: Effect of source of dietary carbohydrate on plasma glucose and insulin responses to mixed meals in subjects with NIDDM.
Diabetes Care
10(4):
395–400,
1987.[Abstract]
- Wolever TM: Metabolic effects of continuous feeding.
Metabolism
39(9):
947–951,
1990.[Medline]
- Trout DL, Behall KM, Osilesi O: Predictions of glycemic index from starchy foods.
Am J Clin Nutr
58:
873–878,
1993.[Abstract/Free Full Text]
- Rasmussen OW, Gregersen S, Dorup J, Hermansen K: Blood glucose and insulin responses to different meals in non-insulin-dependent diabetic subjects of both sexes.
Am J Clin Nutr
56(4):
712–715,
1992.[Abstract/Free Full Text]
- Jenkins DJ, Jenkins AL: Dietary fiber and the glycemic response.
Proc Soc Exp Biol Med
180(3):
422–431,
1985.[Abstract]
- Wolever TM, Vuksan V, Eshuis H, Sqadafora P, Peterson RD, Chao ES, Storey ML, Jenkins DJ: Effect of method of administration of psyllium on glycemic response and carbohydrate digestibility.
J Am Coll Nutr
10(4):
364–371,
1991.[Abstract]
- Sparks SP, Jovanovic-Peterson L, Peterson CM: Blood glucose rise following prenatal vitamins in gestational diabetes.
J Am Coll Nutr
12(5):
543–546,
1993.[Abstract]
- Khan MA, Gannon MC, Nuttall FQ: Glucose appearance rate following protein ingestion in normal subjects.
J Am Coll Nutr
11(6):
701–706,
1992.[Abstract]
- Gulliford MC, Bicknell EJ, Scarpello JH: Differential effect of protein and fat ingestion on blood glucose responses to high- and low-glycemic-index carbohydrates in noninsulin-dependent diabetic subjects.
Am J Clin Nutr
50(4):
773–777,
1989.[Abstract/Free Full Text]
- Burke LM, Collier GR, Hargreaves M: Muscle glycogen stores after prolonged exercise: Effect of the glycemic index of carbohydrate feedings.
J Appl Physiol
75:
1019–1023,
1993.[Abstract/Free Full Text]
- Thomas DE, Brotherhood JR, Brand JC: Carbohydrate feeding before exercise: Effect of glycemic index.
Int J Sports Med
12(2):
180–186,
1991.[Medline]
- Coyle EF: Substrate utilization during exercise in active people.
Am J Clin Nutr
61(4 suppl):
9685–9795,
1995.
TOP
ABSTRACT
INTRODUCTION
